The primary purpose of this project is to describe and evaluate the distribution of cancer occurrence in the United States in terms of age, sex, race, place, and time in order to identify subgroups of the population that offer possibilities for mortality reduction through intervention. Cancer incidence, mortality, and survival rates are analyzed using biostatistical techniques. Data on kidney cancer diagnosed among white residents of Iowa during 1973- 88 were available from the SEER program. Data for 1973-77 were used to compute baseline rates for the major site/type subgroups of kidney cancer within each of the 99 counties in the state. Data for 1978-88 were monitored for increasing incidence over the baseline rates using the sets technique. Alarms signaling an increase in incidence were triggered in 39 counties for renal parenchyma cancer and in 34 counties for renal pelvis cancer. In collaboration with researchers in the Environmental Epidemiology Branch, HHV-6 antibody titters in chronic fatigue syndrome patients were investigated and found to be higher than for controls, but the difference was not statistically significant. In another study for health blood donors from Africa and Asia, HHV-6 reactivity was significantly higher for Ghanaians than for Malaysians or Americans. A retrospective cohort study, done in collaboration with investigators in the Clinical Epidemiology Branch, was designed to detect the effects of cancer and its treatment on childhood patients who survived to adulthood. The occurrence of early menopause among female members of this cohort was determined using survival analysis methodology and compared to the risk among sisters of the survivors. Survivors were found to be at an increased rate for early menopause. The rate was primarily a function of age and the type of treatment received for cancer. In another study based on this cohort, development of the cigarette smoking habit among 1,289 survivors who were nonsmokers when their cancer was diagnosed was compared to that among 1,930 sibling controls. Using conditional maximum likelihood methods, the smoking rate at last follow-up among survivors was found to be 92% of the rate among controls.